Side-by-side benchmark comparison across knowledge, coding, math, and reasoning.
Qwen3.5 397B (Reasoning) is clearly ahead on the aggregate, 82 to 43. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
Qwen3.5 397B (Reasoning)'s sharpest advantage is in mathematics, where it averages 92.1 against 26.4. The single biggest benchmark swing on the page is AIME 2024, 95 to 26.4. GPT-4.1 does hit back in knowledge, so the answer changes if that is the part of the workload you care about most.
Qwen3.5 397B (Reasoning) is the reasoning model in the pair, while GPT-4.1 is not. That usually helps on harder chain-of-thought-heavy tests, but it can also mean more latency and more token spend in real use. GPT-4.1 gives you the larger context window at 1M, compared with 128K for Qwen3.5 397B (Reasoning).
Pick Qwen3.5 397B (Reasoning) if you want the stronger benchmark profile. GPT-4.1 only becomes the better choice if knowledge is the priority or you need the larger 1M context window.
Qwen3.5 397B (Reasoning)
77
GPT-4.1
78.3
Qwen3.5 397B (Reasoning)
67.7
GPT-4.1
54.6
Qwen3.5 397B (Reasoning)
92.1
GPT-4.1
26.4
Qwen3.5 397B (Reasoning)
89
GPT-4.1
87.4
Qwen3.5 397B (Reasoning) is ahead overall, 82 to 43. The biggest single separator in this matchup is AIME 2024, where the scores are 95 and 26.4.
GPT-4.1 has the edge for knowledge tasks in this comparison, averaging 78.3 versus 77. Inside this category, GPQA is the benchmark that creates the most daylight between them.
Qwen3.5 397B (Reasoning) has the edge for coding in this comparison, averaging 67.7 versus 54.6. Inside this category, SWE-bench Verified is the benchmark that creates the most daylight between them.
Qwen3.5 397B (Reasoning) has the edge for math in this comparison, averaging 92.1 versus 26.4. Inside this category, AIME 2024 is the benchmark that creates the most daylight between them.
Qwen3.5 397B (Reasoning) has the edge for instruction following in this comparison, averaging 89 versus 87.4. Inside this category, IFEval is the benchmark that creates the most daylight between them.
Get notified when new models drop, benchmark scores change, or the leaderboard shifts. One email per week.
Free. No spam. Unsubscribe anytime. We only store derived location metadata for consent routing.